TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/mBERT

mBERT

Reported on 20 benchmarks across 5 tasks · 4 papers · 7 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing20 results

  • Natural Language InferenceonFarsTail
    % Test Accuracy· 2020-09-18
    83.38
    SOTA
    FarsTail: A Persian Natural Language Inference DatasetarXiv:2009.08820
  • Cross-LingualonCoNLL Dutch
    F1· 2019-04-19
    77.57
    best: 83.35 (Zero shot mBERT 3)
    SOTA
    Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERTarXiv:1904.09077
  • Cross-LingualonCoNLL German
    F1· 2019-04-19
    69.56
    best: 75.33 (SMTS Multi sim)
    SOTA
    Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERTarXiv:1904.09077
  • Cross-LingualonCoNLL Spanish
    F1· 2019-04-19
    74.96
    best: 79.5 (XLM-R large)
    SOTA
    Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERTarXiv:1904.09077
  • Cross-Lingual TransferonCoNLL Dutch
    F1· 2019-04-19
    77.57
    best: 83.35 (Zero shot mBERT 3)
    SOTA
    Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERTarXiv:1904.09077
  • Cross-Lingual TransferonCoNLL German
    F1· 2019-04-19
    69.56
    best: 75.33 (SMTS Multi sim)
    SOTA
    Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERTarXiv:1904.09077
  • Cross-Lingual TransferonCoNLL Spanish
    F1· 2019-04-19
    74.96
    best: 79.5 (XLM-R large)
    SOTA
    Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERTarXiv:1904.09077
  • Linguistic AcceptabilityonRuCoLA
    MCC· 2022-10-23
    0.15
    best: 0.594 (Ru-RoBERTa+TDA)
    RuCoLA: Russian Corpus of Linguistic AcceptabilityarXiv:2210.12814
  • Linguistic AcceptabilityonItaCoLA
    MCC· 2022-10-23
    0.36
    best: 0.683 (XLM-R + TDA)
    RuCoLA: Russian Corpus of Linguistic AcceptabilityarXiv:2210.12814
  • Subjectivity AnalysisonCzech Subjectivity Dataset
    Accuracy· 2022-04-29
    91.23
    best: 93.56 (XLM-R-Large)
    Czech Dataset for Cross-lingual Subjectivity ClassificationarXiv:2204.13915
  • Cross-LingualonXTREME
    AVG
    59.6
  • Cross-LingualonXTREME
    Question Answering
    53.8
    best: 52.5 (Anonymous5)
  • Cross-LingualonXTREME
    Sentence Retrieval
    47.7
    best: 94.4 (Turing ULR v6)
  • Cross-LingualonXTREME
    Sentence-pair Classification
    73.7
    best: 91 (Turing ULR v6)
  • Cross-LingualonXTREME
    Structured Prediction
    66.3
    best: 90.8 (Creative)
  • Cross-Lingual TransferonXTREME
    AVG
    59.6
  • Cross-Lingual TransferonXTREME
    Question Answering
    53.8
    best: 52.5 (Anonymous5)
  • Cross-Lingual TransferonXTREME
    Sentence Retrieval
    47.7
    best: 94.4 (Turing ULR v6)
  • Cross-Lingual TransferonXTREME
    Sentence-pair Classification
    73.7
    best: 91 (Turing ULR v6)
  • Cross-Lingual TransferonXTREME
    Structured Prediction
    66.3
    best: 90.8 (Creative)